A new algorithm for deducing user search intensities from feedback sessions
N. Sowmiya, R. Chennappan · 2025
When multiple users ask a search engine a vague, general-topic query, it is probable that they have different search objectives. Enhancing search engine relevancy and user experience can be achieved by deducing and analyzing objectives. In this paper, we describe a novel approach to extract user search objectives from query data from search engines. Initially, we suggest a method for grouping the suggested feedback sessions so as to identify different user search intentions for a certain query. Click-through records from users are used to design feedback sessions that accurately represent users’ information demands. Second, we offer a novel method for fabricating fictitious articles that better emulate feedback sessions for clustering. Lastly, “Classified Average Precision (CAP)” is the term we suggest. A novel measure to assess the effectiveness of estimating user search objectives. To support the user search objectives, feedback meetings, fictitious documents, rearranging search results, and classification average precision are all included in the index.